The major aims of the proposed research are to add (1) to our understanding of the development of phonetic competence during early infancy, (2) to our knowledge of the early perceptual capacities of infants and (3) to our knowledge of the processing systems that underlie the perception of speech in mature listeners. More specifically, experiments that use a high amplitude sucking procedure are proposed to investigate whether some of the prerequisites for duplex perception exist within the processing capacity of young infants. These prerequisites include being able to discriminate information in either channel and being able to fuse dichotically presented acoustic information. Additional studies are planned that are relevant to duplex perception, but that will also provide evidence related to the general problem of whether infants perceive the information for speech (in this case formant transitions) in accord with principles that would seem to derive from tacit knowledge of the consequences of production. In another series of experiments with infants, the characteristics of the processing system that mediates the perception of simple visual patterns is to be investigated. The proposed studies, which use a familiarization/novelty preference method, are designed to provide more detailed data on the use of multiple sources of visual information. Of concern is how these processing properties compare to seemingly similar processes in the domain of speech. Finally, two series of experiments are planned with adult listeners. The first series is to provide data on the flow of information during the perception of speech. A modification of the procedure to study duplex perception is to be used to examine the limits on the extent to which information can be used by more than a single processor. The second series of studies investigates the higher-order, decision-related aspects of the speech processing system. These experiments use a priming technique to examine the effects of automatic processes as well as acquired expectations on simple decisions based on the perceived phonetic quality of speech signals. These studies together should serve to enlarge our view of the perceptual systems for speech and provide a base against which deviations in the development of competence for the processing of speech may be evaluated.